Tongue image segmentation method and device

Abstract

The invention provides a tongue image segmentation method and a device. The segmentation method comprises the following steps: acquiring a grayscale tongue image and an HSI colored image according to the original colored image; acquiring a first image based on the grayscale tongue image by utilizing a watershed algorithm; segmenting the HSI colored image based on the data of a channel H of the HSI colored image so as to acquire a second image; segmenting the second image based on the data of a channel I of the HSI colored image so as to acquire a third image; and calculating the set of the first image and the third image, thereby obtaining the tongue segmentation result map. The segmentation method and segmentation device are sensitive to the tongue boundary in the image, and the segmentation result is accurate; and the segmentation speed is high, and the method and device have high timeliness.

Description

technical field [0001] The present invention relates to the field of medical image processing, in particular to a tongue image segmentation method and device. Background technique [0002] There are many medical image segmentation algorithms, commonly used such as threshold segmentation algorithm, watershed segmentation algorithm, active contour model segmentation algorithm, segmentation algorithm based on neural network, etc., but because each algorithm has its own problems, not all methods are applicable to images. Tongue segmentation. The surface of the tongue coating has complex texture features, and the gray scale of the tongue body is similar to that of lips, skin and other organ images, so the boundary of the tongue body is blurred and the boundary information is weak. At the same time, the tongue diagnosis system is ultimately aimed at ordinary patients. The tongue segmentation algorithm needs to be accurate on the one hand, and time-sensitive on the other hand. ...

AI Technical Summary

Problems solved by technology

However, some contours segmented by the local maximum value as a watershed are not the real contours that need to be segmented, causing the real contours to be buried, which is the phenomenon of over-segmentation

The over-segmentation phenomenon of the watershed algorithm will have a serious negative impact on the segmentation effect of the tongue image

[0005] The segmentation method based on Active Contour Models (ACM) is slow to achieve segmentation, and it relies too much on the selection of the initial contour. If the initial contour is not well selected, the algorithm may converge to the local extremum, and the tongue cannot be correctly realized. volume segmentation

In addition, the active contour model method is an iterative algorithm. In the case of a large deviation in the initial contour selection, if you want to obtain a good result, the number of iterations will increase, which increases the calculation time.

Poor timeliness and slow speed are major disadvantages of the active contour model method

[0006] The neural network-based segmentation algorithm has problems such as large amount of calculation, long time consumption, and slow convergence speed, and is not suitable for tongue segmentation in medical images.

Images